Abstract Considerable evidence implicates GABAergic dysfunction in the psychosis spectrum, which includes schizophrenia and bipolar disorder. GABAergic agents treat psychosis spectrum patients, often for anxiety, and the response of catatonic patients to GABA-enhancing benzodiazepines suggests a deeper link between GABA and psychosis. Post-mortem work strongly implicates GABAergic interneurons in these patients, but the links between these findings, clinical phenotype and therapeutics are far from clear. Accordingly, this proposal will undertake a transdiagnostic approach to elaborate linkages between GABA and the negative affective states (NA, e.g. anxiety, stress sensitivity) that are common across this psychosis spectrum. Using a multimodal neuroimaging strategy, magnetic resonance spectroscopy (MRS) with be combined with a pharmaco-fMRI (phfMRI) challenge paradigm measuring blood oxygenation level dependent (BOLD) changes. MRS studies of GABA levels in the psychosis spectrum have yielded inconsistent findings, which this proposal will address by positing two processes: 1) Elevated GABA in early psychosis patients in the rostral medial frontal cortex (rMFC), and 2) Reduced GABA associated with more NA. The investigators will pursue preliminary data showing that NA may be linked to low GABA levels in the rMFC, and when controlling for NA, early psychosis patients have higher GABA in the rMFC. In Aim 1, this study will utilize GABA MRS to assess GABA levels in psychosis spectrum patients, comparing early episode patients (50% not taking antipsychotics) with healthy controls, as well as schizophrenia, schizoaffective and bipolar patients. By probing three distinct voxels in the medial frontal cortex, the investigators will show specific regional effects in the rMFC and demonstrate that antipsychotic medication is associated with reduced GABA levels. In Aim 2, the phfMRI challenge paradigm with a benzodiazepine will be used to measure dynamic GABA function. In published work, the investigators have demonstrated abnormal BOLD response (increased rather than decreased signal) in schizophrenia in the dorsomedial prefrontal cortex (dmPFC). This BOLD response to a GABAergic manipulation is correlated with NA in both patients and controls, consistent with a transdiagnostic dimension. The project will extend these findings across the psychosis spectrum. As the regulation of GABA function plays a critical role in the excitatory/inhibitory (E/I) balance, the phfMRI probe serves as a proxy measure of E/I balance, and it will be used to test the hypothesis that this dynamic measure is associated with NA and GABA concentration measured with MRS. Expected Impact: Successful achievement of study aims will clarify GABAergic dysfunction in the psychosis spectrum, leading to follow-up studies on the role of illness stage in GABA activity and the E/I balance. As NA is a strong, transdiagnostic predictor of functional outcome, the impact of this work will establish potential therapeutic leverage points linking GABAergic treatments with NA as a clinical outcome in the psychosis spectrum.